# easyAI-pro

**Repository Path**: guoliang1983/easyAI-pro

## Basic Information

- **Project Name**: easyAI-pro
- **Description**: No description available
- **Primary Language**: Java
- **License**: Apache-2.0
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 1
- **Forks**: 1
- **Created**: 2026-05-31
- **Last Updated**: 2026-05-31

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# easyAI-pro

> **Disclaimer: This project is an enhanced improvement based on easyAI. Currently in research stage, please be aware of risks for commercial use.**

## Project Introduction

easyAI-pro is a pure Java artificial intelligence algorithm framework with no external dependencies, ready to use out of the box. It includes pre-built image object detection and AI customer service modules, as well as various underlying algorithm tools such as deep learning, machine learning, reinforcement learning, heuristic learning, and matrix operations. Developers can easily learn to develop small-scale models that meet their specific business needs.

### Project Features

- **Zero Dependencies**: Pure Java implementation, no additional environment configuration required
- **Out of the Box**: One-click Maven integration, start using immediately
- **Comprehensive**: Covers machine learning, deep learning, NLP, reinforcement learning, and more
- **Easily Extensible**: Modular design for easy customization based on business needs

## Modules

### Machine Learning

#### Classification Algorithms
| Algorithm | Description |
|-----------|-------------|
| KNN | K-Nearest Neighbors |
| SVM | Support Vector Machine |
| DecisionTree | Decision Tree |
| AdaBoost | Adaptive Boosting |
| XGBoost | Extreme Gradient Boosting |
| LightGBM | Light Gradient Boosting |
| CatBoost | Categorical Boosting |
| LogisticRegression | Logistic Regression |
| GaussianNaiveBayes | Gaussian Naive Bayes |
| MultinomialNaiveBayes | Multinomial Naive Bayes |
| ComplementNaiveBayes | Complement Naive Bayes |
| GradientBoosting | Gradient Boosting Classification |

#### Clustering Algorithms
| Algorithm | Description |
|-----------|-------------|
| KMeans | K-Means Clustering |
| DBSCAN | Density-Based Clustering |
| HDBSCAN | Hierarchical Density Clustering |
| GMM | Gaussian Mixture Model |
| Birch | Hierarchical Clustering |
| MeanShift | Mean Shift Clustering |
| OPTICS | Ordering Points Clustering |
| SpectralClustering | Spectral Clustering |
| HierarchicalClustering | Hierarchical Clustering |

#### Regression Algorithms
| Algorithm | Description |
|-----------|-------------|
| LinearRegression | Linear Regression |
| RidgeRegression | Ridge Regression |
| LassoRegression | Lasso Regression |
| ElasticNetRegression | Elastic Net Regression |
| SVR | Support Vector Regression |
| PolynomialRegression | Polynomial Regression |
| HuberRegression | Huber Regression |
| RANSACRegression | RANSAC Regression |
| QuantileRegression | Quantile Regression |
| DecisionTreeRegression | Decision Tree Regression |

#### Ensemble Learning
| Algorithm | Description |
|-----------|-------------|
| Bagging | Bootstrap Aggregating |
| Stacking | Stacking Ensemble |
| VotingClassifier | Voting Classifier |
| ExtraTrees | Extra Randomized Trees |
| GBM | Gradient Boosting Machine |
| RandomForest | Random Forest |

### Deep Learning

#### Neural Network Basics
- Neuron models
- Activation functions: ReLU, Sigmoid, Tanh, ELU
- Backpropagation algorithm

#### Convolutional Neural Networks
| Model | Description |
|-------|-------------|
| ResNet | Residual Network |
| UNet | Image Segmentation Network |

#### Object Detection
| Model | Description |
|-------|-------------|
| YOLO | Real-time Object Detection |
| FastYolo | Fast YOLO |

#### Generative Models
| Model | Description |
|-------|-------------|
| GAN | Generative Adversarial Network |
| DCGAN | Deep Convolutional GAN |
| BigGAN | Large-scale GAN |
| BigGANHD | High Resolution BigGAN |
| StyleGAN | Style-based GAN |
| StyleGAN3 | StyleGAN Third Generation |
| StyleGANXL | Extra Large StyleGAN |
| CycleGAN | Cycle-Consistent GAN |
| VAE | Variational Autoencoder |
| VQVAE | Vector Quantized VAE |
| VQVAE2 | Vector Quantized VAE V2 |
| BetaVAE | Beta Variational Autoencoder |
| BetaTCVAE | Beta-TC Variational Autoencoder |
| ConditionalVAE | Conditional VAE |
| VAEGAN | VAE-GAN Hybrid Model |

#### Diffusion Models
| Model | Description |
|-------|-------------|
| DiffusionModel | Diffusion Model Base |
| StableDiffusion | Stable Diffusion |
| StableDiffusionV2 | Stable Diffusion V2 |
| DALLE | DALL·E Image Generation |
| DALLE2 | DALL·E 2 |
| MAE | Masked Autoencoder |
| BEiT | BERT Pretraining Image Transformer |

### Natural Language Processing

#### Word Embedding Models
| Model | Description |
|-------|-------------|
| Word2Vec | Word Vector Training |
| GloVe | Global Vectors |
| FastText | Fast Text Vectors |

#### Pretrained Language Models
| Model | Description |
|-------|-------------|
| BERT | Bidirectional Encoder Representations |
| DistilBERT | Distilled BERT |
| RoBERTa | Robustly Optimized BERT |
| ALBERT | A Lite BERT |
| GPT | Generative Pretrained Transformer |
| GPT3 | GPT Third Generation |
| GPT4 | GPT Fourth Generation |
| AutoGPT | Automatic GPT |
| T5 | Text-to-Text Transformer |
| Pegasus | PEGASUS Summarization Model |
| BART | Bidirectional Auto-Regressive Transformer |
| XLNet | Permutation Language Model |
| Llama | LLaMA Large Model |
| LLaMA2 | LLaMA 2 |
| Mistral | Mistral Model |
| BLOOM | BLOOM Large Model |
| Claude | Claude Model |

#### Text Processing
| Feature | Description |
|---------|-------------|
| TF-IDF | Term Frequency-Inverse Document Frequency |
| TextClassification | Text Classification |
| Tokenizer | Tokenizer |
| KeyWord | Keyword Extraction |
| Sentence | Sentence Processing |
| Word | Word Processing |

### Time Series

| Model | Description |
|-------|-------------|
| LSTM | Long Short-Term Memory Network |
| GRU | Gated Recurrent Unit |
| BidirectionalLSTM | Bidirectional LSTM |
| LSTMAutoencoder | LSTM Autoencoder |
| LSTMSeq2Seq | LSTM Sequence to Sequence |
| LSTMVariants | LSTM Variants |
| ARIMA | Autoregressive Integrated Moving Average |
| SARIMA | Seasonal ARIMA |
| Prophet | Prophet Forecasting Model |
| NBEATS | Neural Basis Expansion Analysis |
| Time2Vec | Time Vector Encoding |
| VAR | Vector Autoregression |
| ExponentialSmoothing | Exponential Smoothing |
| TemporalFusionTransformer | Temporal Fusion Transformer |
| TimeSeriesTransformer | Time Series Transformer |
| TransformerTimeSeries | Transformer Time Series Model |

### Reinforcement Learning

| Algorithm | Description |
|-----------|-------------|
| DQN | Deep Q-Network |
| PPO | Proximal Policy Optimization |
| A3C | Asynchronous Advantage Actor-Critic |
| A2C | Advantage Actor-Critic |
| DDPG | Deep Deterministic Policy Gradient |
| SAC | Soft Actor-Critic |
| TD3 | Twin Delayed DDPG |
| SARSA | State-Action-Reward-State-Action |
| IMPALA | Importance Weighted Actor-Learner Architecture |
| IQN | Implicit Quantile Network |
| Rainbow | Rainbow DQN |
| Dreamer | Dreamer Agent |
| QLearning | Q-Learning |
| DynamicProgramming | Dynamic Programming |

### Graph Neural Networks

| Model | Description |
|-------|-------------|
| GCN | Graph Convolutional Network |
| GAT | Graph Attention Network |
| GraphSAGE | Graph Sampling and Aggregation |
| GraphRNN | Graph Recurrent Neural Network |
| GraphTransformer | Graph Transformer |
| GraphBERT | Graph BERT |
| HeterogeneousGNN | Heterogeneous Graph Neural Network |
| GNNExplainer | GNN Explainer |
| Node2Vec | Node Vector Representation |
| Louvain | Louvain Community Detection |

### Attention Mechanisms

| Model | Description |
|-------|-------------|
| ViT | Vision Transformer |
| ViT22B | ViT 22 Billion Parameter Version |
| ViTModel | ViT Model |
| SwinTransformer | Swin Transformer |
| SwinTransformerV2 | Swin Transformer V2 |
| Longformer | Long Document Transformer |

### Self-Supervised Learning

| Method | Description |
|--------|-------------|
| SimCLR | Contrastive Learning |
| BYOL | Bootstrap Your Own Latent |
| MoCoV3 | Momentum Contrast V3 |
| CLIP | Contrastive Language-Image Pretraining |
| MAE | Masked Autoencoder |
| SimSiam | Simple Siamese Network |
| BarlowTwins | Barlow Twins |
| DINOV2 | DINO V2 |

### Anomaly Detection

| Algorithm | Description |
|-----------|-------------|
| IsolationForest | Isolation Forest |
| LOF | Local Outlier Factor |
| LocalOutlierFactor | Local Outlier Factor |
| EllipticEnvelope | Elliptic Envelope |
| OneClassSVM | One-Class SVM |

### Association Rule Mining

| Algorithm | Description |
|-----------|-------------|
| Apriori | Apriori Algorithm |
| FPGrowth | FP-Growth Algorithm |
| Eclat | Eclat Algorithm |

### Dimensionality Reduction & Feature Engineering

#### Dimensionality Reduction Algorithms
| Algorithm | Description |
|-----------|-------------|
| PCA | Principal Component Analysis |
| TSNE | t-SNE Dimensionality Reduction |
| UMAP | Uniform Manifold Approximation |
| Isomap | Isometric Mapping |
| LDA | Linear Discriminant Analysis |
| LLE | Locally Linear Embedding |

#### Feature Selection
| Method | Description |
|--------|-------------|
| RFE | Recursive Feature Elimination |
| MutualInformation | Mutual Information |
| SelectKBest | Select K Best Features |
| SelectPercentile | Select Percentile Features |
| VarianceThreshold | Variance Threshold |
| SelectFromModel | Select From Model |

### Recommender Systems

| Method | Description |
|--------|-------------|
| CollaborativeFiltering | Collaborative Filtering |
| ContentBasedFiltering | Content-Based Filtering |
| MatrixFactorization | Matrix Factorization |

### Other Modules

#### Voice Processing
| Module | Description |
|--------|-------------|
| Whisper | Whisper Speech Recognition |
| WaveFile | Wave File Processing |
| MP3 | MP3 Audio Processing |

#### Probabilistic Models
| Model | Description |
|-------|-------------|
| BayesianNetwork | Bayesian Network |

#### Optimization Algorithms
| Algorithm | Description |
|-----------|-------------|
| PSO | Particle Swarm Optimization |

#### Matrix Operations
| Module | Description |
|--------|-------------|
| Matrix | Matrix Basic Operations |
| CudaMatrix | CUDA Accelerated Matrix |
| SimdMatrix | SIMD Accelerated Matrix |
| SVD | Singular Value Decomposition |
| QRMatrix | QR Decomposition |
| MatrixOperation | Matrix Operations |
| MatrixPool | Matrix Pool |
| AdvancedMatrixOperation | Advanced Matrix Operations |
| MatrixMulAccelerate | Matrix Multiplication Acceleration |

### Tool Modules

| Tool | Description |
|------|-------------|
| AutoML | Automated Machine Learning |
| DataAugmentation | Data Augmentation |
| DataPreprocessing | Data Preprocessing |
| FeatureEngineering | Feature Engineering |
| ModelCompression | Model Compression |
| ModelExporter | Model Export |
| ModelInterpretation | Model Interpretation |
| ModelSelection | Model Selection |
| ModelTrainer | Model Trainer |
| ModelVisualizer | Model Visualizer |
| HyperparameterOptimization | Hyperparameter Optimization |
| AdvancedHyperparameterOptimization | Advanced Hyperparameter Optimization |
| PerformanceMetrics | Performance Metrics |
| ImageTools | Image Tools |
| CreateImageData | Image Data Creation |

## Installation

### Requirements

- JDK 1.8 or higher
- Maven 3.x

### Maven Dependency

```xml
<dependency>
    <groupId>org.dromara.easyai</groupId>
    <artifactId>easyAi</artifactId>
    <version>1.6.1</version>
</dependency>
```

## Quick Start

### KMeans Clustering Example

```java
import org.dromara.easyai.clustering.KMeans;
import org.dromara.easyai.entity.Cluster;

public class KMeansExample {
    public static void main(String[] args) {
        KMeans kMeans = new KMeans();
        kMeans.setK(3);
        List<Cluster> clusters = kMeans.fit(data);
    }
}
```

### LSTM Time Series Prediction Example

```java
import org.dromara.easyai.time_series.LSTM;

public class LSTMExample {
    public static void main(String[] args) {
        LSTM lstm = new LSTM();
        lstm.train(trainData);
        float[] prediction = lstm.predict(inputData);
    }
}
```

### BERT Text Processing Example

```java
import org.dromara.easyai.nlp.BERT;

public class BERTExample {
    public static void main(String[] args) {
        BERT bert = new BERT();
        float[] embeddings = bert.encode("Hello World");
    }
}
```

## Project Structure

```
src/main/java/org/dromara/easyai/
├── anomaly/                 # Anomaly Detection
├── anomaly_detection/        # Anomaly Detection Extensions
├── association/              # Association Rules
├── association_rules/        # Association Rules Extensions
├── attention/               # Attention Mechanisms
├── batchNerve/              # Batch Neural Networks
├── bayesian/                # Bayesian Algorithms
├── classification/           # Classification Algorithms
├── clustering/              # Clustering Algorithms
├── computer_vision/          # Computer Vision
├── config/                  # Configuration Classes
├── conv/                    # Convolution Related
├── deep_learning/           # Deep Learning
├── diffusion/               # Diffusion Models
├── dimensionality/           # Dimensionality Reduction
├── dimensionality_reduction/ # Dimensionality Reduction Extensions
├── ensemble/                # Ensemble Learning
├── entity/                  # Entity Classes
├── feature_selection/       # Feature Selection
├── function/                # Activation Functions
├── gameRobot/               # Game Robots
├── graph/                   # Graph Algorithms
├── graph_neural_network/    # Graph Neural Networks
├── matrixTools/             # Matrix Tools
├── naturalLanguage/         # Natural Language Processing
├── nerveCenter/             # Neural Network Center
├── nerveEntity/             # Neural Network Entities
├── nlp/                     # NLP Models
├── probabilistic/           # Probabilistic Models
├── pso/                     # Particle Swarm Optimization
├── randomForest/            # Random Forest
├── recommend/               # Recommender Systems
├── recommendation/          # Recommender Systems Extensions
├── regression/              # Regression Algorithms
├── regressionForest/        # Regression Forest
├── reinforcement/           # Reinforcement Learning
├── reinforcement_learning/  # Reinforcement Learning Extensions
├── resnet/                  # ResNet Implementation
├── rnnJumpNerveCenter/      # RNN Jump Connection Center
├── rnnJumpNerveEntity/      # RNN Jump Connection Entities
├── rnnNerveCenter/          # RNN Center
├── rnnNerveEntity/          # RNN Entities
├── self_supervised/         # Self-Supervised Learning
├── time_series/             # Time Series
├── tools/                   # Utility Classes
├── transFormer/             # Transformer Implementation
├── unet/                    # UNet Implementation
├── utils/                   # Utility Classes
├── vehicle/                 # Vehicle Related
├── voice/                   # Voice Processing
└── yolo/                    # YOLO Object Detection
```

## Contributing

1. Fork the repository
2. Create a new branch (`git checkout -b feature/AmazingFeature`)
3. Commit your changes (`git commit -m 'Add some AmazingFeature'`)
4. Push to the branch (`git push origin feature/AmazingFeature`)
5. Create a Pull Request

## License

This project is licensed under the Apache License 2.0. See [LICENSE](http://www.apache.org/licenses/LICENSE-2.0.txt)

## Acknowledgments

This project is an enhanced improvement based on [easyAI](https://gitee.com/dromara/easyAi). Special thanks to the original authors.

---

## Comparison with Original easyAI - Enhancement Points

This project has been significantly enhanced based on the original easyAI. Here are the main newly added modules and features:

### 1. Core Module Additions

#### Machine Learning Algorithm Extensions
| Module | New Content |
|--------|------------|
| **Classification** | CatBoost, LightGBM, LogisticRegression, GaussianNB, MultinomialNB, ComplementNB, GradientBoosting |
| **Clustering** | HDBSCAN, SpectralClustering, HierarchicalClustering |
| **Regression** | ElasticNet, Polynomial, Quantile, DecisionTreeRegression |
| **Feature Selection** | MutualInformation, SelectKBest, SelectPercentile, VarianceThreshold, SelectFromModel |
| **Anomaly Detection** | IsolationForest, EllipticEnvelope, LocalOutlierFactor, OneClassSVM |
| **Association Rules** | Eclat |

#### Deep Learning & Generative Models
| Module | New Content |
|--------|------------|
| **GANs** | BigGAN, BigGANHD, StyleGAN, StyleGAN3, StyleGANXL, CycleGAN, DCGAN |
| **VAEs** | VQVAE, VQVAE2, BetaVAE, BetaTCVAE, ConditionalVAE, VAEGAN |
| **Diffusion Models** | StableDiffusion, StableDiffusionV2, DALLE, DALLE2, BEiT |
| **Computer Vision** | ResNet, UNet, ImageProcessing |

#### Natural Language Processing
| Module | New Content |
|--------|------------|
| **Large Models** | GPT-3, GPT-4, LLaMA, LLaMA2, Claude, BLOOM, Mistral, T5, BART, XLNet |
| **Transformer Variants** | ALBERT, DistilBERT, RoBERTa |
| **Language Processing** | FastText, GloVe, TextClassification |

#### Graph Neural Networks
| Module | New Content |
|--------|------------|
| **Core Models** | GAT, GraphSAGE, GraphTransformer, GraphBERT |
| **Extensions** | HeterogeneousGNN, GNNExplainer |

#### Attention Mechanisms & Self-Supervised Learning
| Module | New Content |
|--------|------------|
| **Vision Transformer** | ViT, ViT22B, SwinTransformer, SwinTransformerV2, Longformer |
| **Self-Supervised** | SimCLR, BYOL, MoCoV3, CLIP, MAE, SimSiam, BarlowTwins, DINOV2 |

#### Reinforcement Learning Extensions
| Module | New Content |
|--------|------------|
| **Core Algorithms** | A2C, DDPG, Dreamer, IMPALA, IQN, Rainbow, SAC, TD3, QLearning |

#### Time Series Enhancements
| Module | New Content |
|--------|------------|
| **Deep Learning Models** | BidirectionalLSTM, LSTMAutoencoder, LSTMSeq2Seq, LSTMVariants |
| **Modern Models** | Prophet, NBEATS, TemporalFusionTransformer, TimeSeriesTransformer |

#### Recommender Systems
| Module | New Content |
|--------|------------|
| **Collaborative Filtering** | CollaborativeFiltering, ContentBasedFiltering, MatrixFactorization |

#### Tools & Utility Modules
| Module | New Content |
|--------|------------|
| **ML Tools** | AutoML, DataAugmentation, FeatureEngineering, ModelCompression, ModelExporter, ModelInterpretation |
| **Performance** | ModelVisualizer, PerformanceMetrics, AdvancedHyperparameterOptimization |
| **Probabilistic** | BayesianNetwork |

### 2. Module Enhancements

| Module | Improvements |
|--------|-------------|
| **Transformer** | Added model submodule (CodecBlockModel, MultiSelfAttentionModel, etc.), added selfAttention submodule (SelfAttention, LayNorm, etc.) |
| **NLP** | Added languageCreator submodule |
| **Regression Forest** | Added Forest, LinearRegression, RegressionForest |
| **Matrix Operations** | Added AdvancedMatrixOperation, MatrixMulAccelerate |
| **Utils** | Added DataUtils, MathUtils, ModelUtils |

### 3. Dependency Enhancements

| Dependency | Purpose |
|------------|---------|
| **Apache POI** | Excel file processing |
| **HanLP** | Chinese NLP processing |

### 4. Summary

**Enhancement Scale**: Expanded from ~30 modules in the original to 60+ main modules, with 20+ new algorithm categories.

**Core Advantages**:
- ✅ Comprehensive support for modern deep learning models (GAN series, diffusion models, Transformer series)
- ✅ New graph neural network modules
- ✅ Complete NLP large model support
- ✅ Rich time series prediction capabilities
- ✅ Enhanced reinforcement learning algorithm library
- ✅ Complete toolchain and utility modules
- ✅ Chinese NLP support (HanLP integration)
- ✅ High-performance matrix computation optimization

---

**Note: This project is currently in research stage, please be aware of risks for commercial use.**